Pathogenicity of Aspergillus Clavatus Produced in a Fungal Biofilm Bioreactor Toward Culex Quinquefasciatus (Diptera: Culicidae)

J. Pestic. Sci. 39(3), 1–6 (2014) DOI: 10.1584/jpestics.D14-006

Original Article

Pathogenicity of Aspergillus clavatus produced in a fungal biofilm bioreactor toward Culex quinquefasciatus (Diptera: Culicidae)

Fawrou Seye,1,2,† Thomas Bawin,2,*,† Slimane Boukraa,2 Jean-Yves Zimmer,2 Mady Ndiaye,1 Frank Delvigne3 and Frédéric Francis2

1 Laboratory of Reproductive Biology, Department of Animal Biology, Faculty of Science and Technology, University Cheikh Anta Diop, P.O. Box-5005, Dakar Fann, Senegal 2 Functional and Evolutionary Entomology, Gembloux Agro-Bio Tech, University of Liege, Passage des Déportés 2, P.O. Box-5030 Gembloux, Belgium 3 Bio-Industries/C.W.B.I., Gembloux Agro-Bio Tech, University of Liege, Passage des Déportés 2, P.O. Box B-5030 Gembloux, Belgium (Received January 21, 2014; Accepted May 18, 2014)

Many entomopathogenic fungi have been demonstrated to be potential agents for efficiently controlling mosquito populations. In the present study, we investigated a bioreactor system to produce metabolites and conidia by combining technological advantages of submerged and solid-state fermentations. The efficiency of fungal products was tested toward mosquitoes. Aspergillus clavatus (Eurotiales: Trichocomaceae) was grown by semi-solid-state fermentation in a bioreactor for up to 7 days. Depending on conidial doses (2.5×107, 5×107, 7.5×107, 10×107 and 12.5×107 conidia/mL), mortality ranged from 37.2±15.0 to 86.3±5.0% toward larvae and from 35.8±2.0 to 85.2±1.5% toward adults. The metabolites (10, 20, 40, 60, 80 and 100% v/v) yielded mortality from 23.7±15.0 to 100.0±0.1% toward larvae, and two sprayed volumes (5 and 10 mL) reached 45.5±1.4 and 75.6±2.6% mortality, respectively, toward adults. © Pesticide Science Society of Japan Keywords: biological control, entomopathogenic fungi, Aspergillus clavatus, biofilm bioreactor, solid-state fermentation, submerged fermentation.

efficient against mosquitoes.8,9) The possibility of using Asper- Introduction gillus niger metabolites against Cx. quinquefasciatus, Anopheles Mosquitoes (Diptera: Culicidae) are permanent blood sucking stephensi and Aedes aegypti larvae has also been shown.10) vectors of diseases such as dengue, filariasis, and malaria. Many The development of a reliable and efficient fermentation sys- mosquito species and strains are resistant to insecticides com- tem for large-scale production of insecticidal products is still monly used for their control. Biological agents have become needed. Classical submerged fermentation of filamentous mi- increasingly attractive alternative for mosquito control. Among croorganisms in aqueous suspensions may impair the produc- potential microorganisms, entomopathogenic fungi have been tion of conidia and metabolites of biotechnological interest.11–13) used and have provided most interesting results in several pest By contrast, solid-state fermentation is a promising alternative controls. Conidia are classically applied, but toxic metabolites since conidia and metabolites are generally produced in higher have also been produced in liquid medium and used against quantities when these microorganisms are produced on a solid mosquitoes.1,2) Metarhizium anisopliae and Beauveria bassiana substrate.11–16) For example, Aspergillus oryzae has been reported are promising alternatives to conventional insecticides for elimi- to produce a 500-fold higher yield of recombinant chymosin in nating mosquito vectors.3,4) Recently, many Aspergillus species solid-state fermentation than in submerged fermentation.17) In have commonly been used against mosquitoes.5,6) The larvicidal this context, efficient utilization of agro-industrial residues as effect of Aspergillus flavus and Aspergillus parasiticus metabo- carbon sources has been shown for the mass production of ento- lites for controlling Culex quinquefasciatus has been shown.7) mopathogenic fungi.18–21) However, the absence of free water in- Aspergillus clavatus conidia have also been demonstrated to be duces parameter variations (such as pH, temperature, moisture,

dissolved oxygen or CO2) that are difficult to control. A comple- mentarity method between solid- and liquid-state fermentations † These authors contributed equally to this work. 22) * To whom correspondence should be addressed. has been attempted in sequential culture. E-mail: [email protected] A fungal biofilm reactor combining the technological advan- Published online ●●● ●●, ●●●● tages of submerged fermentation (i.e., free water facilitating the © Pesticide Science Society of Japan control of culture parameters) with the biological characteris- 2 F. Seye et al. Journal of Pesticide Science tics found in solid-state fermentation has been established in the present study. Insecticidal activity of conidia and metabolites secreted by an entomopathogenic A. clavatus strain in this culture system was shown against Cx. quinquefasciatus larvae and adults. Materials and Methods 1. Fungal strain An A. clavatus (Eurotiales: Trichocomaceae) strain isolated from the locust cricket Oedaleus senegalensis (Orthoptera: Acrididae) at the Laboratory of Reproductive Biology of the University Cheikh Anta Diop (Dakar, Senegal) and shown to be pathogenic against Anopheles gambiae, Ae. aegypti and Cx. quinquefasciatus larvae9) was used (accession number MUCL 55275, Belgian Coordinated Collections of Microorganisms, Mycothèque de l’Université Catholique de Louvain (Belgium)). This strain was cultivated on potato dextrose agar (PDA) in Petri dishes and stored at 4°C.

2. Inoculum As a preculture, A. clavatus mycelia were produced in Erlenmey- er conical flasks (500 mL) containing 300 mL of distilled water, Fig. 1. Schematic diagram of a 20-L bioreactor system with metal packing. The packed structure (16 cm height×16 cm diameter) consisted of 1% peptone, 1% yeast extract and 2% glucose sterilized at 120°C a stainless steel cylinder composed of several independent rectangular for 20 min. These cultures were incubated after inoculation for plates surrounded by a broad plate. The cylinder was closed at the bot- 48 hr on a rotary shaker at 140 rpm and 30°C to produce fungal tom by a circular plate. Uniform liquid media were ensured by mixing. pellets. The pump drew the liquid containing the fungal pellets provided from a preculture, and injected them into the packing containing wheat bran as a solid substrate. Biomass immobilization was ensured to avoid liquid vis- 3. Metal structured packing for fungal sporulation cosity. To facilitate pellets fixation and fungal sporulation in a bioreactor, a metal packing was molded. Packing is a cylinder composed of several stainless steel corrugated sheets independent containing wheat bran. The mycelial pellets were then carried of each other, as previously described.23) For packing, 28 metal out with the liquid medium and fixed to the molded system. The rectangular plates (16×16 to 2×16 cm) with a 2-mm mesh were temperature was maintained at 30°C by circulating temperature- cut for assembly. The plates were superimposed symmetrically controlled water. Air was continuously supplied to the bioreac- in pairs following the decreasing width of the rectangle in order tor at 3 L/min. The pH was allowed to vary freely and was re- to form a cylinder approximately 16 cm in diameter. All were corded continuously. Cultivation was carried out for 7 days. then surrounded by a broad plate to form a block of cylindri- cal packages 16 cm in height and 16 cm in diameter. One of the 5. Conidia recovery bases was covered by a circular metal to prevent solid substrates After incubation, the packing that contained fungal conidia from falling into the liquid medium. and biomass was removed, and the liquid medium was stored at −80°C. The packing was washed with an aqueous solution 4. Bioreactor system containing 0.05% Tween 80 (0.05% Tween 80 solution) to detach The fermentation run was carried out in a 20-L bioreactor with the conidia. The washing solution was then filtered to discard 6-L working volume. Wheat bran (200 g) was introduced as a impurities and centrifuged (4°C, 3,000 g, 5 min). The conidia carbon source in the cylinder between the metal plates of the were finally suspended in a 0.05% Tween 80 solution and the packing, which was suspended in the Biolafitte bioreactor (Fig. dose was adjusted to 109 conidia per mL using a hemocytometer 1). Peptone and yeast extract (60 g each) were previously intro- (Thoma®). duced in the total working volume of the reactor containing 6 L of distilled water and 0.05 g/L chloramphenicol as a bacte- 6. Filtration of metabolites riostatic agent. The preparation was autoclaved at 120°C. The Liquid medium from the bioreactor was centrifuged (4°C, preculture was aseptically injected into the bioreactor using a 7,500 g, 30 min) to eliminate impurities. Medium was then fil- syringe. With a peristaltic pump (120 rpm), the culture medium tered with ultrafiltration membranes (Prep/Scale-TFF-1 Car- was continuously stirred (connections made with silicone tubing tridge (Millipore) 5.8 cm in diameter and 15.2 cm long with with an internal diameter of 5 mm) and injected on the packing polyethersulfone membrane filtration (column 1 ft2 cartridge)) Vol. 39, No. 3, 000–6 (2014) Aspergillus metabolites and conidia produced in a biofilm bioreactor 3 to remove unconsumed nutrients from the culture medium. was sprayed on the adults. In parallel, 5 and 10 mL of concen- Final concentrated metabolites (800 mL) with high molecular trated metabolites were sprayed directly on the groups of 20 weight products (MW>100,000) were obtained and stored at adults previously introduced into separate cages and kept for 3 −80°C. days. For control, distilled water was sprayed on the adults. 8.3 Statistical analysis 7. Mosquito rearing Four replicates were performed for each treatment for different Cx. quinquefasciatus (S-Lab) adults were reared in 50-×50- durations with new independent fermentation products. Mortal- ×50-cm cages and fed with a 10% sucrose solution. Blood meal ity was recorded daily for three days and corrected using Ab- was made using the Hemotek feeding system. Larvae were main- bott’s formula.24) Dead larvae were removed, rinsed with sterile tained in distilled water (25-×15-×5-cm containers) in labora- distilled water to eliminate non-attached conidia, and examined tory conditions (25±2°C, 70±5% relative humidity, and 16 : 8 hr under a microscope for fungal infection. Dead larvae and adults (Light : Dark) photoperiod) and fed with fish food (TetraMin®) were incubated in Petri dishes containing wet filter paper to and natural brewer’s yeast tablets (Biover®). monitor the conidial germination. All experiments were done in laboratory conditions with 25±2°C temperature and a 16-hr 8. Bioassays light photoperiod. 8.1 Larval treatment Results First, groups of 20 third-instar larvae were exposed to conidial doses of 2.5×107, 5×107, 7.5×107, 10×107 and 12.5×107 conid- 1. Pathogenicity of conidia on Culex quinquefasciatus ia/mL in separate bottles for 3 days. Control larvae were main- Conidia were effective against Cx. quinquefasciatus larvae and tained in 0.05% Tween 80 solution. Second, larvae were exposed adults. The percentage of larval mortality increased with increas- to serial dilutions (10, 20, 40, 60, 80, and 100% (v/v)) of concen- ing conidial concentration and ranged from 37.2±15.0 to 86.3± trated metabolites with distilled water for 3 days. Control larvae 5.0% for 72 hr of treatment (Table 1). Adult mortality ranged were maintained in distilled water. All larvae were fed with the from 35.8±2.0 to 85.3±1.5% after 72 hr post-inoculation (Table same food as used for rearing. 2). Conidial adhesion to the larval cuticle was not observed. 8.2 Adult treatment Nevertheless, conidia ingested by larvae during bioassays invad- Conidia were formulated with 5% sunflower oil (v/v) in 0.05% ed the larval gut, germinated, and emerged thereafter from dead Tween 80 solution. The same conidial suspension series used larvae (Fig. 2A–C). By contrast, the conidia penetrated the adult against larvae were directly sprayed on a total of 20 three- and mosquito’s cuticle before emerging (Fig. 2D). five-day-old sugar-fed adults that were transferred in 25-×25- ×25-cm mosquito netting cages and kept for 3 days. For control, 2. Toxicity of metabolites on Culex quinquefasciatus a 0.05% Tween 80 solution containing 5% sunflower oil (v/v) Metabolites produced by A. clavatus in a bioreactor showed tox-

Table 1. Mortality of Culex quinquefasciatus larvae treated with Aspergillus clavatus conidia produced in bioreactor*

Conidial concentration (107 conidia/mL) Treatment time 2.5 5 7.5 10 12.5 Cumulative mortality (%)±SD 24 hr 15.8±14.0 17.1±17.0 30.3±20.0 38.2±15.0 60.5±15.0 48 hr 25.0±6.0 35.5±22.0 42.1±30.0 47.4±9.0 77.6±13.0 72 hr 37.2±15.0 55.3±5.0 65.8±9.0 73.8±6.0 86.3±5.0 * Values are average corrected mortality (%)±standard deviation (SD).

Table 2. Mortality of Culex quinquefasciatus adults treated with Aspergillus clavatus conidia produced in bioreactor*

Conidial concentration (107 conidia/mL) Time after inoculation 2.5 5 7.5 10 12.5 Cumulative mortality (%)±SD 24 hr 20.0±3.0 20.0±4.0 24.8±2.0 28.3±8.0 37.5±1.0 48 hr 25.8±5.0 55.7±2.0 50.0±0.2 60.0±3.1 75.7±0.8 72 hr 35.8±2.0 56.5±0.9 65.5±0.7 75.0±14.0 85.3±1.5 * Values are average corrected mortality (%)±standard deviation (SD). 4 F. Seye et al. Journal of Pesticide Science

Fig. 2. Culex quinquefasciatus larvae and adults infected with Aspergillus clavatus cultivated in the bioreactor system. Conidia (A) were observed within the larval gut after 24 hr of treatment. Germ tube (arrow) and filaments within the larval gut (B), and conidial heads (arrow) on the larval cadavers (C) were observed 72 hr after treatment. Conidial formation was observed on the thorax (arrow), legs, and wings of the adults 72 hr post-inoculation (D). co: conidia, m: muscle, h: head, ab: abdomen, th: thorax and: wing.

Table 3. Effect of Aspergillus clavatus metabolites on Culex quinquefasciatus larvae*

Concentration of metabolites (%) Treatment time 10 20 40 60 80 100 Cumulative mortality (%)±SD 24 hr 2.6±3.0 22.4±16.0 27.6±10.0 67.1±8.0 76.3±6.0 93.4±5.0 48 hr 11.8±13.0 26.3±15.0 38.2±15.0 73.7±4.0 78.9±4.0 97.4±3.0 72 hr 23.7±15.0 39.5±27.0 55.3±5.0 76.3±6.0 82.9±2.0 100.0±0.1 * Values are average corrected mortality (%)±standard deviation (SD).

Table 4. Effect ofAspergillus clavatus metabolites on Culex quinquefasciatus adults* Discussion

Sprayed doses of metabolites (mL) The system developed in this work shows the possibility to pro- Time after treatment 5 10 duce simultaneously, with controlled parameters, two kinds of A. clavatus products: conidia and metabolites in a bioreactor. Cumulative mortality (%)±SD The substrate contained in the metal packing allowed for effi- ± ± 24 hr 7.7 0.1 18.5 0.2 cient growth of the fungal biomass, which sporulated. The liquid 48 hr 40.1±2.0 65.3±3.6 media allowed for easy recovery of metabolites secreted by the 72 hr 45.5±1.4 75.6±2.6 fungal biomass. Moreover, the biomass remained confined on * Values are average corrected mortality (%)±standard deviation (SD). the substrate and did not increase the viscosity of the liquid medium. The present process highlights the possibility to control the parameters of metabolites and conidia production for fur- icity against Cx. quinquefasciatus larvae and adults. The mor- ther study. Homogenization of the physical traits in the packed tality increased with increasing metabolite concentration and structure (pH, oxygen, temperature), as well as their impact on ranged from 23.7±15.0 to 100.0±0.1% for larvae after 72 hr conidia and metabolites production, must be investigated. Other post-treatment (Table 3) and 45.5±1.4 to 75.6±2.6% for adults entomopathogenic fungi, such as M. anisopliae and B. bassiana, after 72 hr of treatment (Table 4). could also be cultivated in this bioreactor system for industrial Vol. 39, No. 3, 000–6 (2014) Aspergillus metabolites and conidia produced in a biofilm bioreactor 5

production. Acknowledgements Conidia and metabolites were both found to be effective for controlling Cx. quinquefasciatus larvae and adults. Unexpect- The authors thank the Islamic Development Bank who supported this work via a post-doctoral fellowship to Fawrou Seye for High Technology edly, conidia cultivated on wheat bran were not strongly patho- in Gembloux Agro-Bio Tech (University of Liege). genic toward Cx. quinquefasciatus larvae as compared with approximately the same dose of conidia cultivated on wheat References powder.9) Less virulence may be related to the kind of substrate ) 1 S. S. Mohanty, K. Raghavendra, P. K. Mittal and A. P. Dash: J. Ind. 25) used for fungal production as previously discussed. However, Microbiol. Biotechnol. 35, 1199–1202 (2008). the pathogenicity of the conidia toward adult mosquitoes was 2) G. Singh and S. Prakash: J. Parasitol. Res. 2011, 147373 (2011). not decreased as compared with previous results, where 86.0± 3) E. J. Scholte, K. Ng’habi, J. Kihonda, W. Takken, K. Paaijmans, S. Ab- 1.7% mortality at day 5 was obtained with 7.9×107 conidia/mL.8) dulla, G. F. Killeen and B. G. Knols: Science 308, 1641–1642 (2005). The sunflower oil used for the fungal formulation facilitated the 4) C. K. Kikankie, B. D. Brooke, B. G. Knols, L. L. Koekemoer, M. Far- conidial adhesion on the adult cuticle as Neem oil against Cx. enhorst, R. H. Hunt, M. B. Thomas and M. Coetzee: Malar. J. 9, 71 quinquefasciatus.8) The observation of dead adults and larvae (2010). 5) A. M. de Moraes, G. L. da Costa, M. Z. Barcellos, R. L. de Oliveira confirmed that the mosquito mortality was due to the conidia and P. C. de Oliveira: J. Basic Microbiol. 41, 45–49 (2001). that, respectively, adhered to and penetrated the cuticle or in- 6) N. Abutaha, F. A. Al-Mekhlafi, A. Semlali, A. Baabbad and M. A. vaded the larval gut. In the latter case, larval death may be due Wadaan: Res. J. Biotechnol. 9, 84–89 (2014). to toxins secreted during germination in the gut of larvae, as the 7) M. Govindarajan, A. Jebanesan and D. Reetha: Trop. Biomed. 22, 1–3 digestive tract was previously reported to be a significant infec- (2005). tion site.26,27) 8) F. Seye and M. Ndiaye: Bacteriol. Virusol. Parazitol. Epidemiol. 53, Our results are similar to those of other studies that showed 43–48 (2008). the toxicity of fungal metabolites against mosquitoes.2,28,29) 9) F. Seye, O. Faye, M. Ndiaye, E. Njie and J. M. Afoutou: J. Insect. Sci. 9, A. clavatus metabolites were found to be toxic toward adults 1–7 (2009). 10) N. Soni and S. Prakash: Am. J. Microbiol. 2, 15–20 (2011). and larvae with significant mortality rates after 72hr. Regard- 11) M. Raimbault: Electron. J. Biotechnol. 1, 114–140 (1998). ing the application approach, metabolites could act by contact 12) U. Hölker, M. Höfer and J. Lenz: Appl. Microbiol. Biotechnol. 64, 175– with adults and also by ingestion by larvae. Another study re- 186 (2004). vealed that metabolites of fungi, such as Tolypocladium inflatum 13) J. Barrios-Gonzales: Process Biochem. 47, 175–185 (2012). (Hypocreales: Ophiocordycipitaceae), could attack mosquito 14) B. Dorta, R. J. Ertola and J. Arcas: Enzyme Microb. Technol. 19, 434– larvae and be the main cause of histopathological damages.30) A. 439 (1996). clavatus germination was associated with the secretion of some 15) T. Arzumanov, N. Jenkins and S. Roussos: Process Biochem. 40, 1037– toxins.31–33) These suggest that metabolites produced by the fun- 1042 (2005). gus act on mosquito tissues including the midgut of larvae.9,30,34) 16) G. V. S. B. Prakash, V. Padmaja and R. R. S. Kiran: Bioresour. Technol. These metabolites must be purified and their chemical structure 99, 1530–1537 (2008). 17) K. Tsuchiya, T. Nagashima, Y. Yamamoto, K. Gomi, K. Kitamoto, C. identified. Moreover, the mechanism of the action of these me- Kumagai and G. Tamura: Biosci. Biotechnol. Biochem. 58, 895–899 tabolites must be defined on non-target organisms before their (1994). use for general biocontrol in the field. 18) H. S. D. Santa, O. R. D. Santa, D. Brand, L. P. D. S. Vandenberghe and Conclusion C. R. Soccol: Braz. Arch. Biol. Techn. 48(spe), 51–60 (2005). 19) K. Sahayaraj and S. K. R. Namasivayam: Afr. J. Biotechnol. 7, 1907– The present study shows the high potential utility of fungal pro- 1910 (2008). duction in a liquid phase continuously recirculated on a metal 20) A. A. Assamoi, J. Destain and P. Thonart: Biotechnol. Agron. Soc. En- structured packing in a bioreactor. This new system was devel- viron. 13, 281–294 (2009). oped for fungal growth and metabolite production, combin- 21) F. R. van Breukelen, S. Haemers, R. Wijffels and A. Rinzema: Process Biochem. 46, 751–757 (2011). ing the technological advantages of submerged and solid-state 22) A. D. Burges and N. W. Hussey (eds.): “Microbial Control of Insect fermentations. The process allowed facility in recuperation and Pests and Mite,” Academic Press, London, 1981. purification of conidia (confined on the solid substrate) and me- 23) Q. Zune, D. Soyeurt, D. Toye, M. Ongena, P. Thonart and F. Delvigne: tabolites (contained in the liquid medium). A. clavatus conidia J. Chem. Technol. Biotechnol. 89, 382–390 (2014). and metabolites were virulent and could be a promising alterna- 24) W. S. Abbott: J. Econ. Entomol. 18, 265–267 (1925). tive to chemical control of mosquito larvae and adults. There- 25) F. A. Shah, C. S. Wang and T. M. Butt: FEMS Microbiol. Lett. 251, fore, it is necessary to isolate and show the effect on non-target 259–266 (2005). organisms before use. These results also highlight the possibility 26) C. M. Lacey, L. A. Lacey and D. R. Roberts: J. Invertebr. Pathol. 52, to control the culture parameters for further studies. The ho- 108–118 (1988). 27) G. S. Miranpuri and G. G. Khachatourians: Entomol. Exp. Appl. 59, mogenization of these physical traits in the packed structure, 19–27 (1991). as well as their impact on conidia and metabolites production, 28) N. Vyas, K. K. Dua and S. Prakash: Parasitol. Res. 101, 385–390 must now be investigated. 6 F. Seye et al. Journal of Pesticide Science

(2007). Soc., 415–421 (1944). 29) G. Singh and S. Prakash: Scientific World Journal. doi: 32) T. M. Lopez-Diaz and B. Flannigan: Int. J. Food Microbiol. 35, 129– 10.1100/2012/603984 (2012). 136 (1997). 30) J. Weiser, V. Matha, Z. Zizka and A. Jegorov: Cytobios 69, 179–186 33) M. Sabater-Vilar, R. F. M. Maas, H. De Bosschere, R. Ducatelle and J. (1992). Fink-Gremmels: Mycopathologia 158, 419–426 (2004). 31) F. Bergel, A. L. Morrison, A. R. Moss and H. Rinderknecht: J. Chem. 34) A. Vey and J. M. Quiot: Can. J. Microbiol. 35, 1000–1008 (1989).